General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

Users may download and print one copy of any publication from the public portal for the purpose of private study or research.

You may not further distribute the material or use it for any profit-making activity or commercial gain

You may freely distribute the URL identifying the publication in the public portal If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from orbit.dtu.dk on: Dec 29, 2021

Contamination of the Arctic reflected in microbial metagenomes from the Greenlandice sheetLetter

Hauptmann, Aviaja Zenia Edna Lyberth; Sicheritz-Pontén, Thomas; Cameron, Karen A.; Bælum, Jacob;Plichta, Damian Rafal; Dalgaard, Marlene Danner; Stibal, Marek

Published in:Environmental Research Letters

Link to article, DOI:10.1088/1748-9326/aa7445

Publication date:2017

Document VersionPublisher's PDF, also known as Version of record

Link back to DTU Orbit

Citation (APA):Hauptmann, A. Z. E. L., Sicheritz-Pontén, T., Cameron, K. A., Bælum, J., Plichta, D. R., Dalgaard, M. D., &Stibal, M. (2017). Contamination of the Arctic reflected in microbial metagenomes from the Greenland ice sheet:Letter. Environmental Research Letters, 12(7), [074019]. https://doi.org/10.1088/1748-9326/aa7445

This content has been downloaded from IOPscience. Please scroll down to see the full text.

Download details:

IP Address: 192.38.90.17

This content was downloaded on 21/07/2017 at 11:22

Please note that terms and conditions apply.

Contamination of the Arctic reflected in microbial metagenomes from the Greenland ice sheet

View the table of contents for this issue, or go to the journal homepage for more

2017 Environ. Res. Lett. 12 074019

(http://iopscience.iop.org/1748-9326/12/7/074019)

Home Search Collections Journals About Contact us My IOPscience

You may also be interested in:

A metagenomic snapshot of taxonomic and functional diversity in an alpine glacier cryoconite

ecosystem

Arwyn Edwards, Justin A Pachebat, Martin Swain et al.

Large cryoconite aggregates on a Svalbard glacier support a diverse microbial community including

ammonia-oxidizing archaea

Jakub D Zarsky, Marek Stibal, Andy Hodson et al.

Temporal and spatial patterns of anthropogenic disturbance at McMurdo Station,Antarctica

Mahlon C Kennicutt II, Andrew Klein, Paul Montagna et al.

Microbial nitrogen cycling in Arctic snowpacks

Catherine Larose, Aurélien Dommergue and Timothy M Vogel

Microbial growth under humic-free conditions in a supraglacial stream system on the Cotton Glacier,

Antarctica

Christine M Foreman, Rose M Cory, Cindy E Morris et al.

The diversity and biogeography of the communities of Actinobacteria in the forelands of glaciers at

a continental scale

Binglin Zhang, Xiukun Wu, Gaosen Zhang et al.

Bacteria in the airways of patients with cystic fibrosis are genetically capable of producing VOCs

in breath

Lieuwe D J Bos, Simone Meinardi, Donald Blake et al.

Viral impacts on bacterial communities in Arctic cryoconite

Christopher M Bellas, Alexandre M Anesio, Jon Telling et al.

OPEN ACCESS

RECEIVED

22 February 2017

REVISED

9 May 2017

ACCEPTED FOR PUBLICATION

19 May 2017

PUBLISHED

11 July 2017

Original content fromthis work may be usedunder the terms of theCreative CommonsAttribution 3.0 licence.

Any further distributionof this work mustmaintain attribution tothe author(s) and thetitle of the work, journalcitation and DOI.

Environ. Res. Lett. 12 (2017) 074019 https://doi.org/10.1088/1748-9326/aa7445

LETTER

Contamination of the Arctic reflected in microbial metagenomesfrom the Greenland ice sheet

Aviaja L Hauptmann1,2,3,10, Thomas Sicheritz-Pontén1, Karen A Cameron4,5,6, Jacob Bælum7,Damian R Plichta8, Marlene Dalgaard1 and Marek Stibal9

1 Department of Bio and Health Informatics, The Technical University of Denmark, Kgs. Lyngby, Denmark2 Novo Nordisk Center for Biosustainability, The Technical University of Denmark, Kgs. Lyngby, Denmark3 Ilisimatusarfik, University of Greenland, Nuuk, Greenland4 Department of Geochemistry, Geological Survey of Denmark and Greenland (GEUS), Copenhagen, Denmark5 Center for Permafrost (CENPERM), University of Copenhagen, Copenhagen, Denmark6 Institute of Biological, Environmental & Rural Sciences (IBERS), Aberystwyth University, Penglais, Aberystwyth, United Kingdom7 Chr. Hansen A/S, Hørsholm, Denmark8 Clinical-Microbiomics, Copenhagen, Denmark9 Department of Ecology, Charles University, Prague, Czech Republic10 Author to whom any correspondence should be addressed.

E-mail: alha@uni.gl

Keywords: metagenomics, cryosphere, contamination, Greenland ice sheet, microbial ecology

Supplementary material for this article is available online

AbstractGlobally emitted contaminants accumulate in the Arctic and are stored in the frozenenvironments of the cryosphere. Climate change influences the release of these contaminantsthrough elevated melt rates, resulting in increased contamination locally. Our understanding ofhow biological processes interact with contamination in the Arctic is limited. Through shotgunmetagenomic data and binned genomes from metagenomes we show that microbialcommunities, sampled from multiple surface ice locations on the Greenland ice sheet, have thepotential for resistance to and degradation of contaminants. The microbial potential to degradeanthropogenic contaminants, such as toxic and persistent polychlorinated biphenyls, was found tobe spatially variable and not limited to regions close to human activities. Binned genomesshowed close resemblance to microorganisms isolated from contaminated habitats. These resultsindicate that, from a microbiological perspective, the Greenland ice sheet cannot be seen as apristine environment.

Introduction

Greenland ecosystems are influenced by anthropogeniccontaminants [1], mostly brought by long-rangeatmospheric transport [2]. Industrial emissions haveresulted in the deposition of contaminants on theGreenland ice sheet (GrIS) since the 1850s [3]. Heavymetals mercury and lead, and persistent toxic com-pounds such as dichloro-diphenyl-trichloroethan(DDT) and polychlorinated biphenyls (PCB) arewell-known contaminants in Greenland and have beenfound to bio-accumulate in the food chain [1].Polycyclic aromatic hydrocarbons (PAHs) have beenfoundontheGrIS[4].Ascontaminantsbecometrappedwithin the snow and ice of the cryosphere, these habitatscan be considered as reservoirs of toxic chemicals [2].

© 2017 IOP Publishing Ltd

Climate warming and the resulting increasedmelting of the cryosphere is causing increased releaseof contaminants in the polar regions [2, 5]. Meltingglaciers are becoming secondary sources of contam-inants, including persistent organic pollutants (POPs)such as polychlorinated biphenyls (PCBs) [2, 6].Dynamic interconnections exist between historic andcurrent emissions for POPs, the cryosphere andclimate change, as manifested by the increased releaseof POPs from Alpine glaciers since the 1990s [5].

The knowledge of the role of microbial commu-nities in the fate of Arctic contamination has beenidentified as a knowledge-gap [2]. Cultured isolatesfrom cryoconite (glacier surface debris) communitiesare known to be able to degrade xenobiotics includingPCBs [7]. Cryoconite microbial communities from the

Environ. Res. Lett. 12 (2017) 074019

GrIS incubated under in situ conditions have beenshown to degrade pesticides [8] and recentlymetabolomic data has indicated in situ catabolismof 2,4-dichlorophenoxyacetic acid by cryoconitemicrobiota from the GrIS [9]. Microbial communitiessampled from cryoconite holes (glacier surface meltfeatures containing mineral-microbe aggregates) havealso been shown to harbor genes associated with heavymetal resistance [10]. It has been suggested thatbacteria in snow might lower the risk of bio-availablemercury being incorporated into Arctic food chains[11]. The potential for mercury reduction amongmicrobes in Arctic coastal waters has been found to beable to account for most of the Hg(0) that is producedin high Arctic waters [12]. In the culturable fraction ofBacteria in snow from the high Arctic, 31% of thecommunity were shown to be mercury-resistant,compared to less than 2% in proximate environmentsof freshwater and brine [11]. Mercury concentrationsin snow have been found to correlate with mercuryresistance gene copy numbers of snow microbiota[13]. These results suggest that cryospheric microbialcommunities interact with anthropogenic contami-nants and might be able to remove a fraction of thecontaminants deposited to this environment beforethey are released to downstream ecosystems [8].

In this study we demonstrate the potential forresistance to and degradation of anthropogeniccontamination by microbial communities sampledfrom cryoconite around the GrIS. We assessed thegenetic potential to resist and utilize contaminantsincluding PCBs, polycyclic aromatic hydrocarbons(PAHs), and heavymetalsmercury and lead via shotgunmetagenomic sequencing and binning of bacterialgenomes. To the best of our knowledge this is the largestshotgunmetagenomicdataset of cryoconite to date, andthe first investigation that puts focus on the Greenlandice sheet surface as a contaminated habitat.

Materials and methods

SamplingSampling has been described previously [14, 15]. Inshort, 34 cryoconite samples were collected from theGrIS in five different locations between May andSeptember 2013. Sampling sites included Tasiilaq(TAS) in Southeast Greenland, Qassimiut (QAS) inSouthwest Greenland, Kangerlussuaq (KAN) in WestGreenland, Dark Site (DS) in West Greenland, inlandfrom the Disko Bay area, and Thule (THU) inNorthwest Greenland (figure 1).

Library preparation and sequencingDNA extraction was performed as described previ-ously [14]. DNA shearing and library preparationswere performed according to the

NEXTflex Rapid DNA-Seq Kit, V13.08 (BiooScientific, Austin, TX, USA).

2

Briefly, 250 ng genomic DNA was sheared byCovaris E210 System using 10% duty cycle, intensity of5 cycles per burst of 200 for 300 seconds to create 200bp fragments. The samples were end-repaired andadenylated to produce an A-overhang. Adapterscontaining unique barcodes were ligated on to theDNA. Samples were then purified using bead sizeselection to select for 300–400 bp fragments using theAgercount AMPure XP beads (Beckman Culter,Beverly, MA, USA).

The purified DNA libraries were amplified with adenaturation time of 2 min at 98 °C, followed by 12cycles of denaturation at 98 °C for 30 seconds,annealing at 65 °C for 30 seconds and extension at72 °C for 1 minute according to the protocol. The finalextension was performed at 72 °C for 4 min. DNAquantification was performed using a NanoDrop ND-1000 UV-VIS Spectrophotometer (Thermo FisherScientific, Waltham, MA, USA) and the quality waschecked on an agilent 2100 Bioanalyzer using theBioanalyser DNA High sensitivity kit (AgilentTechnologies, Santa Clara, CA, USA). Librarypreparation was performed by in-house facilities(DTU Multi-Assay Core (DMAC), Technical Univer-sity of Denmark). DNA libraries were mixed into 9pools in equimolar ratios, resulting in 5 samples persequencing lane. Sequencing was performed as a 100bp paired-end run on a HiSeq 2000 (Illumina Int., SanDiego, CA, USA) following the manufacturer’srecommendations. The metagenomic data is availableon the NCBI Sequence Read Archive throughaccession number PRJNA360211.

Demultiplexing and data quality checkThe number of different barcodes in raw sequencingfiles was counted prior to demultiplexing intoindividual sample files. This was done in order toconfirm that all barcodes were represented in roughlyequal numbers in the sequencing output, and to checkfor contamination. Demultiplexed files for each of thesamples from nine lanes of Illumina HiSeq sequencingwere concatenated to obtain one complete sample filein fastq format [16]. Data quality was assessed usingFastQC version 0.11.2 [17] and reads were trimmedfor bad quality bases and Illumina adaptors usingCutadapt [18] through MGmapper version 2.5 [19].Cutadapt settings were used with a minimum phredscore for base quality of 30, and minimum read lengthof 30 bp.

Scaffold assembly and gene callingQuality trimmed fastq files were assembled usingidba_ud assembler version 1.1.1. with pre-correction[20]. Prokaryotic open reading frames were calledusing Prodigal version 2.6.2 [21].

Binning of genomesFiles with genes called by Prodigal for each samplewere concatenated into a gene catalogue and each

80°N 80°W 60°W 40°W 20°W 0°W

75°N

70°N

65°N

60°N

50°W 40°W 20°W

500 km

Figure 1. Map of sampling sites. TAS = Tasiilaq, THU = Thule, DS = Dark Site (one of the darkest 5 km pixels in optical satelliteimagery [15]), KAN = Kangerlussuaq, QAS = Qassimiut. Sample details have been described previously [14,15].

Environ. Res. Lett. 12 (2017) 074019

entry was given a unique header. The gene cataloguewas clustered to reduce homology at 95% identityusing USEARCH package cluster_fast version8.1.1861 [22]. Centroids from the clustering wereused in downstream analyses. For each sample, rawreads were mapped against the indexed genecatalogue using bwa mem version 0.7.12 [23].Samtools version 0.1.18 [24] was used to sort andconvert sam format files from mapping to bamformat files. An in house python script was used tocount the number of reads mapping to genes in thegene catalogue. Read counts were assembled into acount matrix with a perl script (Script 1 supplemen-tary material). The count matrix was rarefied to thedepth of the shallowest sample with a depth of 49 671209 read counts per sample. This was done using theVegan package rrarefy [25] in R Studio version0.99.903 [26]. Genome bins were clustered based on

3

gene co-abundance with the Canopy Clusteringmethod with a maximum canopy distance of 0.1and a maximum merge distance of 0.05 [27]. A totalof 29 bins with correlation to profile > 0.95 andbetween 3000 and 8000 genes were assembled fromcontigs to which binned genes belonged. The qualityof the binned genomes was tested with Checkm [28].The genomes were annotated and further assessed inRAST [29]. Putative genomes are accessible throughRAST guest user, username ‘guest’ and password‘guest’, genome IDs are listed in supplementary table2 available at stacks.iop.org/ERL/12/074019/mmedia.When comparing binned genomes with controlgenomes in RAST the control genomes used wereEscherichia coli 536 publicly available in RAST(control 1), Lactobacillus casei ATCC 334 alsopublicly available in RAST (control 2) as well asBelliella baltica DSM 15883 [30] (control 3).

Environ. Res. Lett. 12 (2017) 074019

Taxonomic compositionMGmapper version 2.5 [31] was used to remove PhiXand map reads from each sample against genomedatabases of Bacteria and Bacteria draft genomes.Results from mapping to the bacterial databases wereused for making stacked barplot of taxonomiccomposition. Quality filters for mapped reads to beincluded in taxonomic composition were; properlypaired reads with alignment scores of >30, minimumcoverage of 80% resulting in minimum identity of86% and a minimum of 10 reads to be mapped to astrain for it to be evaluated as true.

Differential potential for degradation of xenobioticsand heavy metal resistanceGenes in the rarefied count matrix were annotated byblasting against the KEGG database [32] usingDiamond blastx version 0.7.9 [33], resulting in BLASTtabular m8 format sequence IDs. Only genes from theoriginal gene catalogue with blast hits E-value< 1e-50were included in counts. There were several genes inthe count matrix with identical sequence IDs althoughthe count matrix is based on homology-reduced genes.This can be explained by the fact that homologyreduction is sequence based and entries in the countmatrix from different parts of a gene would have thesame annotation but would not be combined in thehomology reduction step (USEARCH cluster_fast).The counts of identical sequence IDs were combined(Script 2 supplementary material) and information ongene names and EC-numbers were added to eachentry in the annotated and combined count matrix(Script 3 supplementary material). Entries in thecount matrix with identical gene names and EC-numbers were combined (Script 4 supplementarymaterial).

The contaminants for which the genetic potentialfor degradation or resistance was assessed wereselected based on known contaminants in the Arcticfood chain [1].

Based on KEGG [32] K0 numbers or gene namesthe number of reads mapping to responsible genes wasextracted from the rarefied count matrix.

For PCB degradation K08689, K08690, K00462 andK10222 were found in the count matrix [32]. For PAHdegradation EC-numbers 1.1.1.256, 1.13.11.3,1.13.11.38, 1.13.11.8, 1.14.12.12, 1.14.12.15, 1.14.12.7,1.14.13.1, 1.14.13.23, 1.2.1.78, 1.3.1.19, 1.3.1.29,1.3.1.49, 1.3.1.53, 1.3.1.64, 3.1.1.35, 4.1.1.55, 4.1.1.69,4.1.2.34 were searched for [32]. For heavy metalresistance genes czc, chr, ncc and mer [34] as well asPbr [35] genes were extracted from the count matrix.

Of the other contaminants listed as important forthe Arctic food chain [1] PFOS were not assessed asevidence suggests that these are not degradedbybacteria[36]. PBDEs were also not assessed as it is still unknownwhich genes are responsible for degradation [37].

For plotting the number of reads mapped todegradation and resistance genes and for comparison

4

between samples, the total number of reads mapped totarget genes were divided by the total number of readsmapped to the 14 genes that have been found to bepresent in most bacterial genomes [38]. This was usedas a measure of an internal constant, since the numberof these genes could be expected not to differ betweenorganisms. In this way, we obtained an approximationof the fraction of degradation and resistance genes persample.

Two metagenomic samples of healthy human gutwere run through the same pipeline as describedabove. These samples were used as negative controls,and in particular, to confirm that genes detected forcontaminant degradation and resistance were notbackground artifacts. Human gut metagenomes weredownloaded from the European Nucleotide Archivestudy PRJEB1220 [27]. We assumed that the healthyhuman guts were non-contaminated. None of thecontaminant related genes were found in the negativecontrols.

Results and discussion

Metagenomic samples were binned into 29 genomesthrough de novo clustering by co-abundance genes(supplementary figure 1). Of the 29 genome bins, 24were successfully annotated in RAST (table 1), and 13bins were complete genomes based on the content oftheir marker genes (table 1; supplementary figure 1).The strain heterogeneity of detected marker genes wasabove 90% for 11 of the 13 bins that were 100%complete (table 1). GC content ranged between 35.4%and 66.2% (mean = 57.7 ± 9.8).

The closest neighbors of resulting putativegenomes were one Cyanobacterium (Nostoc sp.PCC 7120), five Proteobacteria, five Actinobacteria,three Bacteroidetes, two Acidobacteria and twoChloroflexi (table 1). The taxonomy of the closestneighbor is not necessarily the taxonomy of the querygenome but rather an indication that the genomesshare functions. The phyla of the closest neighborswere among the most abundant (>1%) in themetagenomic samples (figure 2). Half of the genomebins belong to families shown to represent >1% ofOTUs in the sampled sites according to a previousstudy based on 16S rRNA gene amplicon sequencing[14]. These were Acetobacteraceae, Acidobacteriaceae,Solibacteraceae, Microbacteriaceae, Chitinophaga-ceae, Sphingobacteriaceae, and Nostocaceae (table 1).16S SSU rRNA genes were located in the binnedgenomes and nucleotide sequences were blastedagainst the NCBI nr/nt database. Top matches arelisted in supplementary table 2. Due to eitherincompleteness or contamination of the binnedgenomes as shown in table 1 and supplementaryfigure 1 some genomes had 0 or several 16S SSU rRNAgenes, which may not be from the same actualgenomes. Additionally, the extracted 16S SSU rRNA

Table 1. Genome bin overview. Closest neighbors are automatically assigned in RAST with the use of a neighbor score. The neighborscore is the number of times that the ‘neighbor’ genome is the top BLAST hit against a candidate gene from the set of ‘unique’ geneswithin the query genome. The closest neighbor does not necessarily indicate the taxonomy of the query genome but indicatessimilarity in assigned functions between the query genome and the closest neighbor.

Bin

ID

Complete-

ness

Contamina-

tion

Strain

heterogeneity

Size (bp) GC

content

(%)

Closest neighbor in RAST

(score)

Phylum Family

81 100 548,45 99,52 23062772 64.5 Acidiphilium cryptum JF-5

(501)

Proteobacteria Acetobacteraceae

59 100 384,08 95,25 15614510 58.8 Acidobacterium sp. MP5ACTX9

(516)

Acidobacteria Acidobacteriaceae

58 100 910,4 99,08 34698014 66 Frankia sp. EuI1c (532) Actinobacteria Frankiaceae

51 100 116,65 79,66 8649830 44.5 Chitinophaga pinensis DSM

2588 (581)

Bacteriodetes Chitinophagaceae

50 100 707,64 97,73 26578570 35.4 Chitinophaga pinensis DSM

2588 (542)

Bacteriodetes Chitinophagaceae

47 100 1305,01 95,36 38823998 65.4 Caulobacter sp. K31 (513) Proteobacteria Caulobacteraceae

45 100 571,82 91,33 26076195 58.7 Acidiphilium cryptum JF-5

(525)

Proteobacteria Acetobacteraceae

41 100 782,94 98,52 33343276 57.9 Acidobacterium sp. MP5ACTX9

(511)

Acidobacteria Acidobacteriaceae

35 100 1051,64 99,63 38873183 65.2 Chloroflexus aggregans DSM

9485 (393)

Chloroflexi Chloroflexaceae

34 100 728,28 97,53 31824964 36.8 Chitinophaga pinensis DSM

2588 (613)

Bacteriodetes Chitinophagaceae

28 100 767,26 52,11 27944005 64.8 Clavibacter michiganensis

subsp. michiganensis NCPPB

382 (526)

Actinobacteria Microbacteriaceae

22 100 651,67 98,87 37385964 49.2 Nostoc sp. PCC 7120 (648) Cyanobacteria Nostocaceae

21 100 1117,12 94,33 48259586 63.3 Streptosporangium roseum

DSM 43021 (503)

Actinobacteria Streptosporangiaceae

77 98,28 300,52 80,03 9563229 63.4 Mucilaginibacter paludis DSM

18603 (838)

Bacteriodetes Sphingobacteriacea

18 98,09 376,32 93,83 23602306 56.8 Ktedonobacter racemifer DSM

44963 (549)

Chloroflexi Ktedonobacteraceae

63 96,55 164,54 90,64 9599541 64.7 Acidiphilium cryptum JF-5

(522)

Proteobacteria Acetobacteraceae

73 95,83 463,02 98,19 18407868 66.2 Acidiphilium cryptum JF-5

(514)

Proteobacteria Acetobacteraceae

48 95,83 348,87 98,45 17680617 65.4 Gluconacetobacter

diazotrophicus PAl 5 (522)

Proteobacteria Acetobacteraceae

29 95,83 668,8 99,58 26933316 63.3 Solibacter usitatus Ellin6076

(501)

Acidobacteria Solibacteraceae

09 95,83 432,39 64,45 36290647 62.4 Leptothrix cholodnii SP-6 (507) Proteobacteria Unclassified

84 95,61 107,35 80,77 5422601 60.7 marine actinobacterium

PHSC20C1 (537)

Actinobacteria Unclassified

30 92,11 75,41 93,02 10047939 47.3 Desulfuromonas acetoxidans

(501)

Proteobacteria Desulforomonadaceae

40 88,95 95,22 6,67 8729345 40.7 Pedobacter heparinus DSM

2366 (583)

Bacteriodetes Sphingobacteriacea

69 67,46 43,03 9,09 3811046 62.9 Leifsonia xyli subsp. xyli str.

CTCB07 (517)

Actinobacteria Microbacteriaceae

Environ. Res. Lett. 12 (2017) 074019

genes are not full length and span different regions ofthe gene, which adds some uncertainty compared totargeted amplicon sequencing of 16S SSU rRNA genes.Therefore the taxonomic indications of the putativegenomes should be regarded with caution.

Mean richness, calculated as CatchAll alphadiversity of samples grouped by area, was previouslyestimated to be between 31 and 116 species per groupbased on 16S rRNA gene sequencing [14]. The 18species found among the binned genomes potentiallymade up between 15.5% and 58% of the community.

5

Organisms isolated from contaminated habitats orknown to be resistant to contaminants were among theclosest neighbors to the binned genomes. The closestneighbor of bin 9, Leptothrix cholodnii SP-6, hasreceived interest as a potential bioremediation organ-ism due to its ability to oxidize iron and manganese,and its potential ability to sequester heavy metals [39].Frankia sp. strain EuI1c, the closest neighbor of bin 58,belongs to a group of organisms known to be lesssensitive to heavy metals such as lead and arsenic [40].This strain in particular has been shown to have

DS

1D

S2

KA

N_P

A2b

KA

N_P

A3a

KA

N_P

B3a

KA

N_P

C1a

KA

N_P

C2a

KA

N_P

C3a

QA

SL1

QA

SL2

TAS

_LA

1aTA

S_L

A1b

TAS

_LA

2aTA

S_L

A2b

TAS

_LA

3aTA

S_L

A3b

TAS

_LB

1aTA

S_L

B1b

TAS

_LB

2aTA

S_L

B2b

TAS

_LB

3aTA

S_L

B3b

TAS

_LC

1aTA

S_L

C1b

TAS

_LC

2aTA

S_L

C2b

TAS

_LC

3aTA

S_L

C3b

TAS

_UA

1aTA

S_U

A1b

TAS

_UA

2aTA

S_U

A2b

THU

1TH

U2

A: Highly abundant Phyla (>1%)

Perc

enta

ge o

f rea

ds m

appe

d to

hig

hly

abun

dant

Phy

la (>

1%)

0

20

40

60

80

AcidobacteriaActinobacteriaArmatimonadetesBacteroidetesCandidatus SaccharibacteriaChlorobiChloroflexiCyanobacteriaDeinococcus ThermusFirmicutesGemmatimonadetesPlanctomycetesProteobacteriaVerrucomicrobia

DS

1D

S2

KA

N_P

A2b

KA

N_P

A3a

KA

N_P

B3a

KA

N_P

C1a

KA

N_P

C2a

KA

N_P

C3a

QA

SL1

QA

SL2

TAS

_LA

1aTA

S_L

A1b

TAS

_LA

2aTA

S_L

A2b

TAS

_LA

3aTA

S_L

A3b

TAS

_LB

1aTA

S_L

B1b

TAS

_LB

2aTA

S_L

B2b

TAS

_LB

3aTA

S_L

B3b

TAS

_LC

1aTA

S_L

C1b

TAS

_LC

2aTA

S_L

C2b

TAS

_LC

3aTA

S_L

C3b

TAS

_UA

1aTA

S_U

A1b

TAS

_UA

2aTA

S_U

A2b

THU

1TH

U2

B: Low abundant Phyla (<=1%)

Perc

enta

ge o

f rea

ds m

appe

d to

low

abu

ndan

t Phy

la (<

=1%

)

0.00

0.05

0.10

0.15

Aquificaecandidate division NC10Candidatus CampbellbacteriaCandidatus PeregrinibacteriaChlamydiaeChrysiogenetesDeferribacteresElusimicrobiaFibrobacteresFusobacteriaIgnavibacteriaeNitrospiraeSpirochaetesSynergistetesTenericutesThermotogaeUnclassified

Figure 2. Taxonomic composition across samples of (a) most abundant Phyla (>1%) and (b) low abundant Phyla (�1%).

Environ. Res. Lett. 12 (2017) 074019

cellular mechanisms for increased copper tolerance[41]. The closest neighbor of bin 47, Caulobacter sp.K31, was isolated from chlorophenol-contaminatedgroundwater and has the ability to tolerate copper andchlorophenols [42]. Four of the bin genomes (bins 81,45, 73 and 63) were closest neighbors to Acidiphiliumcryptum JF-5, which was isolated from acidic coalmine lake in eastern Germany and was shown to beable to reduce Fe(III) [43]. These binned putativegenomes indicate that cryoconite holes on the GrIScontain organisms with the ability to survive andutilize the presence of long-distance transportedcontaminants.

The percentages of genes in pathways (RASTactivesubsystems) related to resistance to and degradation ofcontaminants were compared among the genome binslisted above, and with the genomes of knownorganisms that are not from cryoconite, namelyEscherichia coli, Lactobaccillus casei, and Belliellabaltica, an isolate from the Baltic Sea [30], as controls.

There were a number of active subsystemsassociated with contaminants, for which somebinned genomes had a comparably high percentageof genes (figure 3). Bin 9 (nearest neighbor Leptothrixcholodnii SP-6) had a notably high percentage ofgenes for zinc resistance (figure 3(a)). Bin 73 (nearestneighbor Acidiphilium cryptum JF-5) had highpercentages of genes in mercuric reductase, coppertolerance, and resistance to chromium compoundsas well as metabolism of aromatic compounds(figures 3(b)–(e)). In general, the binned genomeshad higher percentages of genes in metabolism ofaromatic compounds when compared to the controls(figure 3(e)).

6

Samples from all the sites had reads that mappedto PCB degradation genes (figure 4(a)). Counts ofreads mapped to PCB degradation genes were variablebetween samples taken from each region. The highestnumber of reads that mapped to PCB degradationgenes were found among the TAS samples fromTasiilaq in Southeast Greenland, ranging from 0(TAS-U-A1b) to 145 (TAS-L-C3b) reads per sample(normalized counts 2.8� 10�4± 3.8� 10�4) (figure 4(a), supplementary table 1). Two TAS samples hadmore than twice as high counts of PCB degradationgenes compared to all of the other samples (1.70 �10�3 and 1.05 � 10�3 PCB degradation reads/housekeeping reads).

For PAH degradation genes, the KAN samplesfrom Kangerlussuaq had the most reads mapped tothese genes followed by the THU (Thule) samples(figure 4(b)). The relatively high abundance of PAHdegradation genes in the KAN samples contrasts theobservations from the other contaminant degradationgenes, which were found to be in relatively lowabundances at KAN (figures 4(a), (c)–(e)). PAHs aregenerated primarily through the incomplete combus-tion of organic materials such as oil, petrol and wood[44]. The PAH content in snow from the Kangerlus-suaq region was assessed previously under thehypothesis that the proximity of the site to the largestairport of Greenland would result in a PAHcontamination [4]. While PAH contamination wasfound to be below detection limit in the study, theresults were inconsistent with previous results andwell below the concentrations found in near-bycryosphere environments [4, 45, 46]. The relativelyhigh abundance of reads mapping to PAH degradation

BIN

63B

IN73

BIN

45B

IN81

BIN

47B

IN58

BIN

09C

ontro

l1C

ontro

l2C

ontro

l3

A: Zinc resistance

Perc

enta

ge o

f gen

es in

Act

ive

Sub

syst

ems

0.0

0.2

0.4

0.6

0.8

1.0

1.2

BIN

63B

IN73

BIN

45B

IN81

BIN

47B

IN58

BIN

09C

ontro

l1C

ontro

l2C

ontro

l3

B: Mercuric reductase

Perc

enta

ge o

f gen

es in

Act

ive

Sub

syst

ems

0.00

0.05

0.10

0.15

BIN

63B

IN73

BIN

45B

IN81

BIN

47B

IN58

BIN

09C

ontro

l1C

ontro

l2C

ontro

l3

C: Copper tolerance

Perc

enta

ge o

f gen

es in

Act

ive

Sub

syst

ems

0.00

0.05

0.10

0.15

0.20

0.25

BIN

63B

IN73

BIN

45B

IN81

BIN

47B

IN58

BIN

09C

ontro

l1C

ontro

l2C

ontro

l3

D: Resistance to chromium compounds

Perc

enta

ge o

f gen

es in

Act

ive

Sub

syst

ems

0.00

0.02

0.04

0.06

0.08

0.10

0.12

BIN

63B

IN73

BIN

45B

IN81

BIN

47B

IN58

BIN

09C

ontro

l1C

ontro

l2C

ontro

l3

E: Metabolism of aromatic compounds

Perc

enta

ge o

f gen

es in

Act

ive

Sub

syst

ems

0.0

0.5

1.0

1.5

2.0

2.5

Potential for resistance to contaminants across genomes

Figure 3. Percentage of genes in contamination-associated active subsystems of the total number of genes in active subsystems withineach genome (count data from RAST). Control genomes were organisms known from other habitats: Escherichia coli 536 (control 1),Lactobacillus casei ATCC (control 2) as well as Belliella baltica DSM 15883 (control 3).

0.00

000.

0005

0.00

100.

0015

A: Normalized PCB degradation gene read count

DS KAN QAS TAS THU

0.01

00.

015

0.02

00.

025

0.03

00.

035

B: Normalized PAH degradation gene read count

DS KAN QAS TAS THU

0.06

0.08

0.10

0.12

0.14

C: Normalized Hg resistance gene read count

DS KAN QAS TAS THU

0e+0

01e

042e

043e

044e

045e

046e

04

D: Normalized Pb resistance gene read count

DS KAN QAS TAS THU8000

10000

12000

14000

16000

18000

E: Heavy metal resistance gene read count

DS

1D

S2

KA

N_P

A2b

KA

N_P

A3a

KA

N_P

B3a

KA

N_P

C1a

KA

N_P

C2a

KA

N_P

C3a

QA

SL1

QA

SL2

TAS

_LA

1aTA

S_L

A1b

TAS

_LA

2aTA

S_L

A2b

TAS

_LA

3aTA

S_L

A3b

TAS

_LB

1aTA

S_L

B1b

TAS

_LB

2aTA

S_L

B2b

TAS

_LB

3aTA

S_L

B3b

TAS

_LC

1aTA

S_L

C1b

TAS

_LC

2aTA

S_L

C2b

TAS

_LC

3aTA

S_L

C3b

TAS

_UA

1aTA

S_U

A1b

TAS

_UA

2aTA

S_U

A2b

THU

1TH

U2

Figure 4. Read counts normalized to counts of reads mapped to housekeeping genes for reads mapped to (a) PCB degradation genes,(b) PAH degradation genes (c) Mercury resistance genes, (d) Lead resistance genes and (e) general heavy metal resistance genes. Notethat counts in panel E have not been normalized to housekeeping genes.

Environ. Res. Lett. 12 (2017) 074019

7

Environ. Res. Lett. 12 (2017) 074019

genes in Kangerlussuaq and Thule samples might beexplained by the proximity of both of these sites tolarge aviation centers (Kangerlussuaq InternationalAirport, and Thule Air Base respectively).

The number of reads that mapped to mercuryresistance genes ranged from 4644 (KAN-P-A2b) to10283 (TAS-L-B1b) (unnormalized counts 7671 ±1586.5). The highest normalized read count ofmercury resistance genes was found in the TASsamples (figure 4(c)). The two QAS and the two THUsamples had similar counts, which were intermediatebetween the high TAS samples and the lower countsin the DS and KAN samples (figure 4(c), supplemen-tary table 1). Mercury concentrations have beenmeasured in the KAN, QAS, TAS and THU regions[4]. Concentrations were lower in snow fromKangerlussuaq (KAN) and higher in the South(QAS) and Southeast (TAS) (supplementary figure 2(a)), which parallels with read count data (figure 4(c)). Inuit from Tasiilaq in Southeast Greenland havealso been shown to have higher contamination withPCBs and mercury in recent years [1]. In contrast,reads for DDT degradation were not found in any ofthe samples.

Lead resistance read counts were found to havevery high numbers in one of the DS samples in WestGreenland and no reads in KAN and QAS samples inSouthwest and South, respectively. The number ofreads that mapped to lead resistance genes rangedbetween 0 to 71 (unnormalized counts 3.18 ± 12.32).The DS-1 sample had almost six times as many readsmapped to lead resistance genes as the other samples(figure 4(d)). In general, the KAN samples had lowernumbers of reads mapped to the combined set ofheavy metal resistance genes (figure 4(e)). Leadconcentrations on the GrIS have been measured tobe slightly higher in the Thule (THU) region(supplementary figure 2(b)) [4]. Read counts of leadresistance genes (figure 4(d)) do not seem to followpreviously measured lead concentrations (supplemen-tary figure 2(b)).

Inuit lead contamination has been shown to besimilar across Greenland with slightly lower contami-nation inTasiilaq inSoutheastGreenland[1], incontrastto other contaminants (figures 4(a)–(c), (e)), and inresemblance to the observations for reads mapped tolead resistance genes in cryoconite (figure 4(d)).

Further research is needed to clarify whether GrISsupraglacial microbial communities may serve asindicators for first entry of contaminants into theArctic food chain, which would have implications forthe release of contaminants into the food chain as aresult of climate warming. This notion is not unlikelyknowing that the copy number of resistance genescorrelates with contaminant concentrations in snow[13], and melting of the cryosphere has been shown torelease contaminants into the downstream ecosystems[2, 5, 6].

8

Our results show that microbial communities onthe GrIS have the potential for resistance to anddegradation of contaminants known to be transportedto the Arctic.

The potential for resistance and degradation wasfound to be spatially variable, yet present at all sitesaround the GrIS and not limited to single samplesites or samples in close proximity to human activity.While the potential was present at all sites this studyis geographically limited to the GrIS. Decreasingcosts of metagenomic analyses and the expectedincrease in the availability of metagenomic datasetsfrom supraglacial habitats will allow the comparisonacross the terrestrial cryosphere. Future researchmight reveal whether there is a broader trend in thedegradation of xenobiotics within the terrestrialcryosphere.

As a negative control the same assessment wascarried out in two metagenomes from healthy humangut samples. The human gut metagenomes had noreads that could be mapped to any of the genesincluded in the cryoconite assessment above. Theseresults suggest that the deposition of contaminantsonto the GrIS may influence the supraglacial microbialcommunity composition and function. Consequently,the GrIS should not be considered a pristineenvironment.

Previous work has shown that gene copy numbersof resistance genes correlate with contaminantconcentrations in snow [13], which suggests thatthe presence of contaminant-related genes inmicrobial communities on the GrIS are a responseto the presence of the relevant contaminants.Metabolomic data from the GrIS indicate in situcatabolism of 2,4-dichlorophenoxyacetic acid bycryoconite microbiota [9]. However, no measure-ments of the contaminants tested in the current studyexist from the sampled sites, and it is not clearwhether or not the genes are expressed. Results fromthe binned genomes are based on active subsystemsonly, which does indicate that the genes are parts offunctional systems in the organisms. Further researchis needed to establish the link between contaminantdeposition on the GrIS and the functional diversity ofthe present microbes.

Contamination of supraglacial environments isconstantly occurring and contaminants from pastdepositions are locked into the ice at certain depths[47]. Concentrations of bioavailable contaminants incryoconite holes may be dependent on the historicdeposition events of the exposed ice surfaces. In thisrespect, thousands of years old contamination mightbecome available to current cryosphere communities.It has been established that current contaminations inhigher trophic levels are largely caused by past events[48]. This might also hold true for microbialcommunities, particularly in supraglacial habitats,where contaminants are locked in the ice.

Environ. Res. Lett. 12 (2017) 074019

Conclusions

The metagenome of microbial communities incryoconite collected on the Greenland ice sheet showsthe potential for resistance to and degradation ofcontaminants, including polychlorinated biphenyls(PCB), polycyclic aromatic hydrocarbons (PAHs), andthe heavy metals mercury and lead. The presence ofgenes for contaminant resistance and degradation wasspatially variable but present in all samples across theice sheet. Binned genomes showed close resemblanceto the genomes isolated from contaminated habitats.Since the genetic potential of contaminant resistanceand degradation usually indicates the presence of therelevant contaminants, the Greenland ice sheet shouldnot be considered a pristine environment, and moreattention should be paid to the potential release ofanthropogenic contaminants from this fast-changingenvironment.

Acknowledgments

This work was partly supported by the Novo NordiskFoundation Center for Biosustainability, the Centerfor Permafrost (CENPERM) Center no 100 from theDanish National Research Foundation (DNRF100) aswell as Dudo and Povl Brandt’s Foundation. This workwas also supported by a Villum Young InvestigatorProgramme grant VKR 023121 to MS. KC is a SêrCYMRU II COFUND fellow who is also supported bythe Sêr Cymru National Research Network for LowCarbon, Energy and Environment.

References

[1] AMAP 2015 AMAP Assessment 2015 Human Health in theArctic (Oslo, Norway) 165 pp

[2] Grannas A M et al 2013 The role of the global cryospherein the fate of organic contaminants Atmos. Chem. Phys. 133271–305

[3] McConnell J R et al 2007 20th-century industrial blackcarbon emissions altered arctic climate forcing Science 3171381–4

[4] Jacobsen C S et al 2012 Global contamination of theGreenlandic icecap with organic contaminants—distributionof contaminants and studies of the microbiological andphotochemical degradation

[5] Bogdal C, Nikolic D, Luthi M P, Schenker U, Scheringer Mand Hungerbuhler K 2010 Release of legacy pollutants frommelting glaciers: model evidence and conceptualunderstanding Environ. Sci. Technol. 44 4063–9

[6] Bogdal C, Schmid P, Zennegg M, Anselmetti F S,Scheringer M and Hungerbuhler K 2009 Blast from thepast: melting glaciers as a relevant source for persistentorganic pollutants Environ. Sci. Technol. 43 8173–7

[7] Margesin R, Zacke G and Schinner F 2002 Characterizationof heterotrophic microorganisms in alpine glaciercryoconite Arctic 34 88–93

[8] Stibal M, Bælum J, Holben W E, Sebastian R, Jensen Aand Jacobsen C S 2012 Microbial degradation of2,4-dichlorophenoxyacetic acid on the greenland ice sheetAppl. Environ. Microbiol. 78 5070–6

9

[9] Cook J M et al 2016 Metabolome-mediated biocryomorphicevolution promotes carbon fixation in Greenlandiccryoconite holes Environ. Microbiol. 18 4674–86

[10] Edwards A et al 2013 A metagenomic snapshot oftaxonomic and functional diversity in an alpine glaciercryoconite ecosystem Environ. Res. Lett. 8 35003

[11] Møller A K, Barkay T, Al-Soud W A, Sørensen S J, Skov Hand Kroer N 2010 Diversity and characterization ofmercury-resistant bacteria in snow, freshwater and sea-icebrine from the high arctic FEMS Microbiol Ecol. 75390–401

[12] Poulain A J et al 2007 Potential for mercury reduction bymicrobes in the high arctic Appl. Environ. Microbiol. 732230–8

[13] Larose C et al 2013 Interactions between snow chemistry,mercury inputs and microbial population dynamics in anarctic snowpack PLoS One. 8 e79972

[14] Cameron K A, Stibal M, Zarsky J D, Gözdereliler E,Schostag M and Jacobsen C S 2016 Supraglacial bacterialcommunity structures vary across the Greenland ice sheetFEMS Microbiol Ecol. 92 fiv164

[15] Stibal M et al 2015 Microbial abundance in surface ice onthe Greenland ice sheet Front Microbiol. 6 225

[16] Cock P J A, Fields C J, Goto N, Heuer M L and Rice P M2010 The Sanger FASTQ file format for sequences withquality scores, and the solexa/illumina FASTQ variantsNucleic Acids Res. 38 1767–71

[17] Patel R K and Jain M 2012 NGS QC toolkit: a toolkit forquality control of next generation sequencing data PLoSOne 7 e30619

[18] Martin M 2011 Cutadapt removes adapter sequences fromhigh-throughput sequencing reads EMBnet.journal 1710–2

[19] Petersen T N, Lukjancenko O and Christen M 2017MGmapper: reference based mapping and taxonomyannotation of metagenomics sequence reads PLoS One 12e0176469

[20] Peng Y, Leung H C M, Yiu S M and Chin F Y L 2012Sequence analysis IDBA-UD: a de novo assembler forsingle-cell and metagenomic sequencing data with highlyuneven depth Bioinformatics 28 1420–8

[21] Hyatt D, Chen G, Locascio P F, Land M L, Larimer F Wand Hauser L J 2010 Prodigal: prokaryotic gene recognitionand translation initiation site identification BMCBioinformatics 11 119

[22] Edgar R C 2010 Search and clustering orders of magnitudefaster than BLAST Bioinformatics 26 2460–1

[23] Li H 2013 Aligning sequence reads, clone sequences andassembly contigs with BWA-MEM

[24] Li H et al 2009 The sequence alignment/map format andSAMtools Bioinformatics 25 2078–9

[25] Oksanen J, Blanchet F G, Kindt R, Legendre P, Minchin PR and O’Hara R B 2013 vegan: Community EcologyPackage. R package version 2.2-1 (http://CRAN.R-project.org/package=vegan)

[26] RStudioTeam 2015 RStudio (Boston, MA: IntegratedDevelopment for R. RStudio, Inc.)

[27] Nielsen H B et al 2014 Identification and assembly ofgenomes and genetic elements in complex metagenomicsamples without using reference genomes Nat. Biotechnol.32 822–32

[28] Parks D H, Imelfort M, Skennerton C T, Hugenholtz P andTyson G W 2015 CheckM: assessing the quality ofmicrobial genomes recovered from isolates, single cells, andmetagenomes Genome Res. 25 1043–55

[29] Aziz R K et al 2008 The RAST server: rapid annotationsusing subsystems technology BMC Genomics 9 75

[30] Brettar I, Christen R and Höfle M G 2004 Belliella balticagen. nov., sp. nov., a novel marine bacterium of thecytophaga-flavobacterium-bacteroides group isolated fromsurface water of the central baltic sea IJSEM 5465–70

Environ. Res. Lett. 12 (2017) 074019

[31] Nordahl Petersen T, Lukjancenko O, Thomsen M C F,Sperotto M M, Lund O, Aarestrup Frank M and Sicheritz-Pontén T 2016 MGmapper: reference based mapping andtaxonomy annotation of metagenomics sequence reads PLoSONE 12 e0176469

[32] Kanehisa M, Furumichi M, Tanabe M, Sato Y andMorishima K 2016 KEGG: new perspectives on genomes,pathways, diseases and drugs Nucleic Acids Res. 45 D353–61

[33] Buchfink B, Xie C and Huson D H 2015 Fast and sensitiveprotein alignment using DIAMOND Nat. Methods 12 59–63

[34] Abou-Shanab R A I, Van Berkum P and Angle J S 2007Heavy metal resistance and genotypic analysis of metalresistance genes in gram-positive and gram-negativebacteria present in ni-rich serpentine soil and in therhizosphere of alyssum murale Chemosphere 68 360–7

[35] Jaroslawiecka A and Piotrowska-seget Z 2014 Leadresistance in micro-organisms Microbiology 160 12–25

[36] Ochoa-Herrera V, Field J A, Luna-velasco A and Sierra-Alvarez R 2016 Microbial toxicity and biodegradability ofperfluorooctane sulfonate (PFOS) and shorter chainperfluoroalkyl and polyfluoroalkyl substances (PFASs)Environ. Sci. 18 1236–46

[37] Xia X 2013 Microbial degradation of polybrominateddiphenyl ethers: current and future Biorem. Biodegrad. 4e130

[38] Neto M R, Galvão C W, Cruz L M, Silva L C, Garcia J Rand Etto R M 2015 Multiple gene sequence analysis usinggenes of the bacterial DNA repair pathway Braz. Arch. Biol.Techn. 58 421–30

[39] Spring S, Kampfer P, Ludwig W and Schleifer K 1996Polyphasic characterization of the genus leptothrix: newdescriptions of leptothrix mobilis sp. nov. and leptothrixdiscophora sp. nov. nom. rev. and emended description ofleptothrix cholodnii emend Syst. Appl. Microbiol. 19 634–43

10

[40] Richards J W, Krumholz G D, Chval M S and Tisa L S2002 Heavy metal resistance patterns of Frankia strainsAppl. Environ. Microbiol. 68 923–7

[41] Rehan M, Furnholm T and Finethy R H 2014 Coppertolerance in Frankia sp. strain EuI1c involves surfacebinding and copper transport Environ. Biotechnol. 988005–15

[42] Ash K, Brown T, Watford T, Scott L E, Stephens C and ElyB 2014 A comparison of the caulobacter NA1000 and K31genomes reveals extensive genome rearrangements anddifferences in metabolic potential Open Biol. 4 140128

[43] Kusel K, Dorsch T, Acker G and Stackebrandt E 1999Microbial reduction of Fe(III) in acidic sediments: isolationof acidiphilium cryptum JF-5 capable of coupling thereduction of Fe(III) to the oxidation of glucose Appl.Environ. Microbiol. 65 3633–40

[44] Abdel-Shafy H I and Mansour M S M 2016 A review onpolycyclic aromatic hydrocarbons: Source, environmentalimpact, effect on human health and remediation Egypt. J.Pet. 25 107–23

[45] Peters A J, Gregor D J, Teixeira C F, Jones N P and SpencerC 1995 The recent depositional trend of polycyclic aromatichydrocarbon to the agassiz ice cap, Ellesmere Island,Canada Sci. Total Environ. 160 167–79

[46] Masclet P, Hoyau V, Jaffrezo J L and Cachier H 2000Polycyclic aromatic hydrocar- bon deposition on the icesheet of greenland. part 1: superficial snow Atmos. Environ.34 3195–207

[47] Stern G A et al 2012 How does climate change influencearctic mercury? Sci. Total Environ. J. 414 22–42

[48] Bjerregaard P, Sloth H, Nielsen N O and Dewailly E 2013Population surveys in Greenland 1993–2009: temporal trendof PCBs and pesticides in the general Inuit population byage and urbanisation Sci. Total Environ. 454–455 283–8